We will develop a proof of principle model for a Drinking Behavioral Control System (DBCS) that will explain the how individuals transition from problem drinking to moderate, non-hazardous, or social drinking. In order to accomplish this aim we will use a dynamic systems modeling approach to model mechanisms of behavior change initiation (MOCBI) for early stage problem drinkers seeking treatment to reduce, but not quit drinking. We will use data from two existing treatment studies of problem drinking to accomplish this task. We will use these results to develop control system simulation that can inform efforts to develop adaptive treatment algorithms for use in future treatment studies. We will also, adapt state-of-the-art cognitive science and neuroscience paradigms related to cognitive control to develop a mechanisms based multi-level (explicit cognitions, implicit cognitions, neurocircuitry) model of self-regulation of problem drinking. During the final year of the project, we will integrate these two areas of investigation to develop a final model and delineate next steps for model testing and validation.